The principal goals of this project are the development of algorithms that allow one to make the best use of NMR data to determine solution structures of biomolecules, to assess in a systematic fashion their accuracy and precision, and to explore the extent to which dynamical information can be extracted from NMR data. This will involve the following components: Updated refinement methods. Refinement models will be developed that use modern protein and nucleic acid force fields in combination with generalized Born or explicit solvation models, and which incorporate conformational disorder through the "locally enhanced sampling" model that uses multiple copies of portions of the macromolecule. Studies on protein and nucleic acid dynamics. Long-time scale molecular dynamics simulations will be used to model NMR relaxation, with attention paid to anisotropic tumbling, to the correlation between internal and overall motions, and to conformational disorder. This will include an analysis of contributions from internal motions to chemical shift anisotropy (CSA) relaxation and to CSA-dipolar cross-correlated relaxation. Slower, microsecond to millisecond motions uncovered by relaxation dispersion experiments will be studied using novel models to identify minor conformers and to estimate their rates of interconversion with other conformations. Initial applications will be to protein G, ribonuclease A, and dihydrofolate reducatase. Nuclear magnetic resonance (NMR) spectroscopy provides a powerful tool for probing the properties of proteins and nucleic acids under conditions like those in living cells. The project uses computational tools to help gain the most information from NMR, promoting our understanding of basic biochemical processes that underlie both healthy and diseased cells. Two of the proteins studied here (ribonuclease and dihydrofolate reductase) are important targets for cancer chemotherapy.